The compounds of the invention are structurally related to buprenorphine. But, unlike buprenorphine, the compounds are mu opioid receptor (MOPr) antagonists that do not also have significant MOPr agonist activity. They share buprenorphine's antagonism at kappa opioid receptors (KOPr) and may also have activity at the NOP/ORL-1 receptor.
Orvinols
The orvinols are a group of ring-C bridged epoxymorphinan compounds of general structure (1) which were originally synthesised by Bentley and co-workers and developed by Reckitt and Colman (Lewis et al, 1971). The most studied members of the series are the extremely potent opiate analgesic and animal immobilising agent etorphine (1a), the very potent opiate antagonist diprenorphine (1b) and the clinical analgesic and treatment for opiate abuse, buprenorphine (1c).

1a: R═H, R1═R2=Me, R3═n-Pr, R4═H
1b: R═H, R1=cyclopropylmethyl, R2═R3=Me, R4═H
1c: R═H, R1=cyclopropylmethyl, R2=Me, R3=t-Bu, R4═H
1d: R═R2═R3═R4═H, R1=cyclopropylmethyl
1e: R═R2═R4═H, R1=cyclopropylmethyl, R3=Me
1f: R═R3═R4═R1=cyclopropylmethyl, R2=Me
The orvinols and derivatives of structure 1 are described in a series of UK Patents GB902659, GB925723, GB937214, GB969263 and GB1136214, which discuss their synthesis and potential therapeutic uses as analgesics, antitussives and, in some cases, antagonists of narcotic drugs. The patent applications do not describe compounds in which R4 is anything other than hydrogen. Further, the stereochemistry about C20 was not specified. The inventors have recognised that the stereochemistry about C20 is in fact of utmost importance and the diastereoisomers of structure 1 in which R2 and R3 are interchanged have very different pharmacological profiles.
In their publications, the inventors have described how the structure-activity relationships in the orvinols (1, R═H) and thevinois (1, R=Me) are conventional with respect to substituents on the basic nitrogen atom and at C3, the phenolic position (Lewis and Husbands, 2004), but that the effect of the nature of the substituents and the stereochemistry of C20 is of even greater significance (Lewis and Husbands, 2004). In particular, the inventors have found that there is only a very limited range of structures which have a lack of opioid agonist activity and are therefore essentially opioid antagonists. Prior to the present invention the only compounds of structure 1 to have been shown to lack agonist activity at opioid receptors were the N-cyclopropylmethyl (N-CPM) primary and secondary alcohols (1d, 1e, 1f) (Lewis, 1973). Thus the discovery of novel orvinols with C20 aryl substituents that lack opioid agonist activity was both unexpected and desirable. Though compounds having structure 1 in which R2 is aryl and R4 is H are within the generic scope of GB969,263 and GB1,136,214, such compounds are not exemplified within those patents.
Orvinols and thevinois of structure 1 where R2 and R3 are interchangeably phenyl and methyl groups, R4 is H and R1 includes cyclopropylmethyl, allyl, dimethylallyl and propargyl were disclosed by Marton et al (Marton et al, 1997) without any definition of opioid activity.
Orvinols and thevinois of structure 1, where R4 is methyl, have not previously been disclosed, in addition, the inventors have found that the key Diels-Alder adduct from which the novel compounds, having R4 as methyl, are prepared in the current invention could not be synthesized by the standard methodology which had previously allowed the standard orvinols, having R4 as H, to be prepared (Lewis et al. 1971).
Buprenorphine in the Treatment of Drug Abuse
Buprenorphine displays a unique and complex pharmacology derived from the manner in which it binds to opioid receptors. Like the opiate analgesics and the antagonists naltrexone, naloxone and nalmefene, buprenorphine's primary actions are at MOPr, but it is neither an opiate agonist like morphine, nor an antagonist like naltrexone. It is classified as a partial MOPr agonist having the characteristics of both an agonist or an antagonist depending on the circumstances. As an MOPr partial agonist buprenorphine shows a ceiling to all the effects associated with MOPr agonism including importantly the potentially lethal effect of respiratory depression (in overdosage) and addiction liability. The latter is also favourably affected by the kinetics of buprenorphine's MOPr binding which has irreversible characteristics. The very slow dissociation of buprenorphine from MOPr is responsible, at least in part, for its long duration of action and the mildness of the abstinence effects when drug is withdrawn following chronic administration. The shape of the dose-response curves for buprenorphine's MOPr agonist effects is uniquely an inverted U-shape. This means that high doses have lesser MOPr agonist effects than intermediate doses which produce peak effects. This applies to respiratory depression, thus further contributing to the drug's extremely favourable acute safety profile and physical dependence liability. The very limited MOPr agonist activity of buprenorphine at high doses is complementary to the predominant MOPr antagonist activity at these doses.
The unique pharmacological profile of buprenorphine at MOPr was recognised when it was approved for the treatment of opiate abuse and dependence as an agent for detoxification and maintenance. But buprenorphine is also unique among drugs having significant MOPr agonist activity in having high affinity but no efficacy at KOPr and delta opioid receptors (DOPr), thus having only antagonist activity at these receptors. Not only does the lack of any KOPr agonist activity mean that buprenorphine avoids KOPr agonist side effects particularly dysphoria and diuresis, but KOPr antagonism is important for its potential use in treatment of substance abuse disorders other than opiate abuse. Thus KOPr antagonism contributes to its ability to inhibit cocaine self-administration in rhesus monkeys (Mello et al, 1995) and in concurrent opiate and cocaine addicts (Montoya et al, 2004). Preclinical studies also support the hypothesis that KOPr antagonists may be of utility against cocaine. The KOPr antagonists norBNI and JDTic have been shown to block stress-induced potentiation of cocaine place preference (McLaughlin et al, 2003) and to block footshock-induced reinstatement of cocaine self-administration behaviour (Beardsley et al, 2005; Redila and Chavkin, 2008). KOPr antagonists have also been shown, in rats, to selectively attenuate ethanol-dependent self-administration while leaving nondependent ethanol self-administration unaffected (Walker and Koob, 2008). This appears consistent with earlier findings of a decrease in alcohol self-administration in KOR knockout mice (Kovacs et al, 2005).
In addition to its binding to the three classical opioid receptors buprenorphine also binds as a partial agonist to the ORL-1 receptor. This receptor has a high degree of amino acid sequence homology with the classical opioid receptors, but traditional opioids, including the opium alkaloids and the antagonists naloxone, naltrexone and nalmefene have low affinity for the ORL-1 receptor. The endogenous ligand for ORL-1 receptors, orphanin FQ has been shown to inhibit the actions of cocaine (Korlinska et al, 2002) so that buprenorphine's ORL-1 activity could be associated with a similar effect. Gorelick (2007) suggested that since it is high doses of buprenorphine that significantly reduce cocaine use in patients who are both opiate and cocaine dependent (Montoya et al, 2004), it is ORL-1 receptor activation which is important for this effect. Buprenorphine at high doses has also been shown to inhibit ethanol self administration in rats, an effect that was prevented by a selective ORL-1 antagonist (Ciccocioppo et al, 2007).
In order to unmask the KOPr antagonist (and NOP/ORL-1 agonist) activity, the MOPr agonist effect of buprenorphine must be nullified. This was achieved by combining sublingual buprenorphine (4 mg/day) with oral naltrexone (50 mg/day) in a study in detoxified opiate addicts. After 12 weeks the combination-treated group showed a lower level of urines positive for opiates and cocaine metabolites than the comparison group on naltrexone (50 mg) (Gerra et al, 2006). This result, which indicated a positive contribution from buprenorphine's KOPr antagonist effect, confirmed an earlier study by Rothman et al (2000) using the same dosing regime, but without a naltrexone comparison group.
The use of naltrexone, a predominantly MOR antagonist, for relapse prevention of alcohol and opioid dependence is approved in a number of countries. There are mixed reports on the effectiveness of oral naltrexone in the treatment of opioid dependence, such that a recent Cochrane review did not find sufficient evidence to unequivocally support its use (Minozzi, 2006). More recently, sustained-release naltrexone has become available and the limited high quality data available suggests this does have advantages over oral naltrexone, including significantly higher rates of retention in treatment (Lobmaier, 2008; Comer et al, 2006). Evidence for the efficacy of naltrexone in treating alcohol dependence is stronger. It has been shown to be effective over both the short and medium term in preventing relapse, particularly when combined with psychosocial treatment (Srisurapanont and Jarusuraisin, 2008). Interestingly, the positive effect in preventing relapse to alcohol is maintained in individuals with dual cocaine/alcohol dependence or abuse (Srisurapanont and Jarusuraisin, 2008), supporting a role for MOR antagonism in reducing alcohol intake in the polydrug using community.
The case is thus made for a potent KOPr antagonist and MOPr antagonist with ORL-1 receptor agonist activity, i.e. buprenorphine with MOPr partial agonism replaced by MOPr antagonism, as a treatment for abuse of a wide spectrum of substances both individually and collectively.